House Prices Predictions using Optimized Ensemble

This is a complete solution of the House Prices - Advanced Regression Techniques competition in Kaggle. The data set has 79 explanatory variables describing (almost) every aspect of residential homes in Ames, Iowa.

Goal

The aim of the competiton is to predict with some a ccuracy the price of each house depending on it features.

Metric

The Competiton submissions are evaluated on Root-Mean-Squared-Error (RMSE) between the logarithm of the predicted value and the logarithm of the observed sales price.

Feature Explination

The data set contains 79 features, here is a breif explination for some of them:

• SalePrice: the property's sale price in dollars. This is the target variable that you're trying to predict.
• MSSubClass: The building class.
• MSZoning: The general zoning classification.
• LotFrontage: Linear feet of street connected to property.
• LotArea: Lot size in square feet.
• Street: Type of road access.
• Alley: Type of alley access.

Using Ensemble for Predictions

The model used is a Regression Ensemble model that contains 4 sub-models. Those models were the best performing models, that were combined to form the Ensemble model, that utilizes all of them to make a prediction about the price of a house. The 4 sub-models are:

• Support Vector Regression
• Ridge Regression
• LGBM Regressor
• Bagging Regressor

Conclusion

A RMSE Score of 0.12448 was achieved by the model.